CN105025493A - Adaptive resource partitioning in a wireless communication network - Google Patents

Abstract

Techniques for performing adaptive resource partitioning are described. In one design, a node computes local metrics for different possible actions related to resource partitioning to allocate available resources to a set of nodes that includes the node. Each possible action is associated with a set of resource usage profiles for the set of nodes. The node sends the computed local metrics to at least one neighbor node in the set of nodes. The node also receives local metrics for the possible actions from the neighbor node(s). The node determines overall metrics for the possible actions based on the computed local metrics and the received local metrics. The node then determines allocation of the available resources to the set of nodes based on the overall metrics. For example, the node may select the action with the best overall metric and may utilize the available resources based on a resource usage profile for the selected action.

Description

Adaptive resource segmentation in cordless communication network

The divisional application that the application is the applying date is on March 19th, 2010, application number is the application for a patent for invention of 201080012038.0.

This application claims the non-provisional United States application No.61/161 being entitled as " UTILITY-BASEDRESOURCE COORDINATION FOR HETEROGENEOUS NETWORKS " submitted on March 19th, 2009, the priority of 646, it is transferred to its assignee and by reference to being incorporated to herein.

Technical field

Disclosure file population relates to communication, more specifically, relates to the technology for support of wireless communication.

Background technology

Cordless communication network is widely deployed to provide various Content of Communication, such as voice, video, grouped data, message transmission, broadcast etc.These wireless networks can be the multi-access networks can being supported multiple user by shared available network resource.The example of this multi-access network comprises code division multiple access (CDMA) network, time division multiple access (TDMA) network, frequency division multiple access (FDMA) network, orthogonal FDMA (OFDMA) network and Single Carrier Frequency Division Multiple Access (SC-FDMA) network.

Cordless communication network can comprise multiple base station, and it can support the communication of multiple user's set (UE).UE can communicate with base station with up link via down link.Down link (or forward link) refers to the communication link from base station to UE, and up link (or reverse link) refers to the communication link from UE to base station.

Base station can send data to UE on the uplink and/or can receive data from UE on uplink.On the uplink, the transmission from base station can observe the interference caused by the transmission from neighbor base station.On uplink, can observe by from the interference caused with the transmission that neighbor base station carries out other UE communicated from the transmission of UE.For both down link and up link, the interference caused by interference base station and interference UE can make performance reduce.Wish to alleviate interference to improve performance.

Summary of the invention

This document describes the technology for performing adaptive resource segmentation in the wireless network.Division of resources refers to process available resources being distributed to node.Node can be base station, relaying or some other entities.For adaptive resource segmentation, available resources dynamically can be distributed to node in the mode that can realize superperformance.

In a design, adaptive resource segmentation can be performed in a distributed way by each node in a group node.In a design, a given node in this group node can calculate the Local Metric about the multiple possible operations relevant to division of resources, so that available resources distribute to this group node.Each possible operation can use profile (profile) to be associated with of this group node group resource.Each resource uses profile can represent the use that be allowed to of a specific node to available resources, such as, and the list of transmit power spectral density (PSD) level be allowed to of available resources.This node can send at least one adjacent node in this group node the Local Metric calculated.This node can also from the Local Metric of this at least one adjacent node reception about the plurality of possible operation.The overall tolerance that this node can be determined about the plurality of possible operation according to calculated Local Metric and the Local Metric received.This node can determine that according to the overall tolerance about the plurality of possible operation the available resources for this group node are distributed subsequently.In a design, this node can select one of them possible operation according to the overall tolerance about these possible operations, such as, selects the possible operation with best overall tolerance.This node subsequently can according to be associated with selected operation and the resource that can be used for this node uses profile to use available resources.Such as, this node can use profile according to the resource of this node, dispatches the transfer of data of at least one UE in available resources.

Multiple scheme of the present invention and feature are below described in more detail.

Accompanying drawing explanation

Fig. 1 shows cordless communication network.

Fig. 2 shows effective group of exemplary UE and the adjacent groups of node.

Fig. 3 shows the process for performing adaptive resource segmentation.

Fig. 4 shows the wireless network with adaptive resource segmentation.

Fig. 5 shows for supporting the process communicated.

Fig. 6 shows for supporting the device communicated.

Fig. 7 shows the process for performing adaptive resource segmentation.

Fig. 8 shows the process of the communication for UE.

Fig. 9 shows the device of the communication for UE.

Figure 10 shows the block diagram of base station and UE.

Embodiment

Technology described herein may be used for various wireless communication network, such as CDMA, TDMA, FDMA, OFDMA, SC-FDMA and other network.Term " network " and " system " usually use interchangeably.Cdma network can realize such as radiotechnics such as universal terrestrial radio access (UTRA), cdma2000 etc.UTRA comprises other variant of wideband CDMA (W-CDMA) and CDMA.Cdma2000 covers IS-2000, IS-95 and IS-856 standard.TDMA network can realize the radiotechnics of such as global system for mobile communications (GSM).OFDMA network can realize such as evolution UTRA (E-UTRA), Ultra-Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20,etc. radiotechnics.UTRA and E-UTRA is a part of Universal Mobile Telecommunications System (UMTS).3GPP Long Term Evolution (LTE) and senior LTE (LTE-A) are the UMTS redactions using E-UTRA, and it uses OFDMA on the uplink and uses SC-FDMA on uplink.UTRA, E-UTRA, UMTS, LTE, LTE-A and GSM is described in the document of the tissue of " third generation partner program " (3GPP) by name.Cdma2000 and UMB is described in the document of the tissue of " third generation partner program 2 " (3GPP2) by name.Technology described herein may be used for above-mentioned wireless network and radiotechnics and other wireless network and radiotechnics.

Fig. 1 shows cordless communication network 100, and it can comprise multiple base station 110 and other network entity.Base station can be the entity communicated with UE, and also can be called the Node B (eNB), access point etc. of node, Node B, evolution.Each base station can provide the communication overlay to specific geographical area.In 3 gpp, according to the context environmental using term, term " community " can refer to the overlay area of base station and/or serve the base station sub-system of this overlay area.In 3GPP2, term " sector " or " cell-sector " can refer to the overlay area of base station and/or serve the base station sub-system of this overlay area.For the sake of simplicity, the 3GPP concept of " community " is employed in describing herein.

Base station can provide the communication overlay of the community to macrocell, picocell, Femto cell and/or other type.Macrocell can cover relatively large geographic area (such as, radius several kilometers), and the UE by having service order can be allowed to carry out untethered access.Picocell can cover relatively little geographic area, and the UE by having service order can be allowed to carry out untethered access.Femto cell can cover relatively little geographic area (such as, family), and can allow to carry out limited access by the UE be associated with this Femto cell (UE such as, in closed user group (CSG)).In the example shown in Fig. 1, wireless network 100 comprises macro base station 110a and 110b for macrocell, for femto base station 110c and 110e of picocell, with for the femto/Home eNodeB 110d of Femto cell.

Wireless network 100 can also comprise relaying.Relaying can be receive transfer of data from upstream entity (such as, base station or UE), and sends the entity of this transfer of data to downstream entities (such as, UE or base station).Relaying also can be the UE of the transmission for other UE of relay forwarding.Relaying also can be called node, stands, relay station, relay base station etc.

Wireless network 100 can be heterogeneous network, and it comprises dissimilar base station, such as, and macro base station, femto base station, femto base station, relaying etc.These dissimilar base stations can have different transmit power level, different overlay areas and the Different Effects to the interference in wireless network 100.Such as, macro base station can have high transmit power level (such as, 20 watts or 43dBm), femto base station can have lower transmit power level (such as, 2 watts or 33dBm), femto base station can have low transmit power level (such as, 0.2 watt or 23dBm).Dissimilar base station can belong to the different capacity grade with different maximum transmit power level.

Network controller 130 can be coupled to one group of base station, and can provide coordination and control for these base stations.Network controller 130 can communicate with base station 110 via backhaul.Base station 110 can also communicate with one another via backhaul.

UE 120 can be dispersed in whole wireless network 100, and each UE can be fixing or movement.UE also can be called station, terminal, mobile radio station, subscriber unit etc.UE can be cell phone, personal digital assistant (PDA), radio modem, Wireless Telecom Equipment, handheld device, kneetop computer, cordless telephone, wireless local loop (WLL) are stood.UE can with the communications such as base station, relaying, other UE.

UE can be positioned at the covering of one or more base station.In a design, single base station can be selected on down link and up link, to serve UE simultaneously.In another design, in down link and up link, each can select a base station to serve UE.For these two designs, serving BS can be selected according to one or more standards of such as maximum geometry (geometry), minimal path loss, ceiling capacity/jamming effectiveness, maximum user throughput etc.Geometry is relevant with received signal quality, and received signal quality can pass through carrier to thermal noise ratio (CoT), signal to noise ratio (SNR), signal and noise and interference ratio (SINR), Carrier interference ratio (C/I) etc. and quantize.Make that energy/jamming effectiveness is maximum can be realized: (i) makes emitted energy needed for every bit minimum, or (ii) makes the reception interfering energy of the per unit of the useful signal energy received minimum.Partly (ii) can be maximum corresponding to the ratio made between the channel gain of expection receiving node and the channel gain summation being all disturbed node.Partly (ii) can be equal to and make the path loss of up link minimum, but may be different for down link.Make user throughput maximum possible to consider various factors, such as, the load (such as, the quantity of the UE that this base station is current served) of base station, distribute to the stock number of base station, the available backhaul capacity of base station etc.

Wireless network can support the one group of resource that can be used for transmitting.Available resources can define according to time or frequency or Time And Frequency both or some other standards.Such as, available resources can correspond to different frequency sub-bands or different time intertexture or different time-frequency chunks etc.Time-interleavedly can comprise evenly spaced time slot, such as, every S time slot, wherein, S can be any integer value.Can be whole wireless network definition available resources.

Base station in wireless network can use available resources in many ways.In an arrangement, each base station can use whole available resources to transmit.This scheme can cause some base stations to obtain poor performance.Such as, the femto base station 110d in Fig. 1 can be positioned near macro base station 110a and 110b, and the transmission from femto base station 110d can observe the high interference from macro base station 110a and 110b.In another scheme, according to fixed resource segmentation, available resources can be distributed to each base station.Each base station can use subsequently its Resources allocation transmit.This scheme can make each base station obtain good performance on its Resources allocation.But some base stations can be assigned with the resource more than required resource, and some other base stations can need resource more more than distributed resource, and this can cause the sub-optimal performance of wireless network.

In an arrangement, adaptive resource segmentation can be performed dynamically available resources are distributed to node, thus good performance can be realized.Division of resources also can be called Resourse Distribute, resource coordination etc.For adaptive resource segmentation on the uplink, can by the list of the transmitting PSD level of specifying can be used in available resources by this node for each node, to each peer distribution available resources.Can to make the maximum mode of utility function (utility function) to perform adaptive resource segmentation.Adaptive resource segmentation is split different from fixing or static resource, and fixing or static resource segmentation can be a fixed subset of each peer distribution available resources.

In a design, mode can perform adaptive resource segmentation in a centralized.In this design, a designated entities can receive the relevant information of UE and node, calculates the tolerance being used for division of resources, and selects best resource to split according to calculated tolerance.In another design, adaptive resource can be performed in a distributed way by a group node and split.In this design, each node can calculate specific tolerance, and can exchange with adjacent node and measure.This metric calculation and exchange execution one can be taken turns or taken turns more.Each node can be determined subsequently and select to provide the division of resources of optimum performance.

Table 1 lists the one group of assembly that can be used for adaptive resource segmentation.

Table 1

assembly	describe
		effective group (Active Set)	for the group node that given UE t retains, it is represented as AS (t).
adjacent groups (Neighbor Set)	for the group node that given node p retains, it is represented as NS (p).
		resource	time and/or the frequency resource of node can be distributed to.
launch PSD level	one group of transmitting PSD level of any given resource can be used for by node.
		utility function	for quantizing the function of the performance of different possible division of resources.

In a design, effective group can be retained for each UE, and it can be determined based on the pilot measurement undertaken by UE and/or the pilot measurement undertaken by node.Effective group of a given UE t can comprise the node with following characteristics: the signal that (i) observes on the uplink for UE t or interference have very important impact, and/or (ii) receives very important signal or interference from UE t on uplink.Effective group also can be called interference management group, candidate set etc.

In a design, can according to CoT effective group as the UE t that gives a definition:

$\mathrm{AS}\left(t\right)=\left\{q\right|\frac{P\left(q\right)\·G(q,t)}{N_0}>{\mathrm{CoT}}_{\min }\}$Equation (1)

Wherein, P (q) is the transmitting PSD of the pilot tone from node q,

G (q, t) is the channel gain between node q and UE t,

N₀the environmental interference and thermal noise that are observed by UE t, and

CoT_minit is the CoT threshold value for selecting to be included in the node in this effective group.

Equation (1) is if represent that the CoT of a given node q is greater than CoT_min, then this node q just can be included in effective group of UE t.Can according to the transmitting PSD of the pilot tone from node q, channel gain between node q and UE t and N₀determine the CoT of node q.Pilot tone can be low leading (LRP) that reuses, or can at low transmitted on resources of reusing thus the location reference signals that can be detected a long way off.Pilot tone also can be pilot tone or the reference signal of certain other type.

Also effective group of UE t can otherwise be defined.Such as, substituting or supplementing as received signal quality, can select node according to received signal strength and/or other standards.Effective group can be limited, to reduce the computation complexity of adaptive resource segmentation.In a design, effective group can be limited to N number of node, and wherein, N can be the numerical value be applicable to arbitrarily.So effective group can comprise and nearly N number ofly to have more than CoT_minthe strongest node of CoT.

In a design, can retain adjacent groups for each node, it can comprise the node participating in adaptive resource segmentation.The adjacent groups of a given node p can comprise the adjacent node with following characteristics: (i) affects the UE served by node p, or (ii) has the UE being subject to node p and affecting.In a design, can the adjacent groups of defined node p as follows:

equation (2)

Wherein, S (t) is the service node of UE t.

Equation (2) is if represent that the given node q of (i) is in effective group of the UE served by node p, or (ii) node q is a service node making node p be included in the UE in its effective group, then node q can be included in the adjacent groups of node p.Thus the adjacent groups of each node can be defined according to effective group of UE and service node thereof.Also otherwise adjacent groups can be defined.Each node according to effective group of the UE by this node serve and the information from adjacent node, can determine its adjacent node.Adjacent groups can be limited to reduce the computation complexity of adaptive resource segmentation.

Fig. 2 shows the exemplary adjacent groups of the node in exemplary effective group and the Fig. 1 of UE in Fig. 1.In the round parentheses being close to each UE, show effective group of this UE in Fig. 2, wherein there is underscore service node/base station.Such as, effective group of UE 1 bem1, M2}, it is meant to this effective group and comprises service node M1 and adjacent node M2.In the square brackets being close to each node, the adjacent groups of this node is shown in Fig. 2.Such as, the adjacent groups of node M 1 is [M2, P1, P2, F1], comprises macro base station M2, femto base station P1 and P2 and femto base station F1.

In a design, can be that each node definition one group launches PSD level, this group transmitting PSD level can comprise whole transmitting PSD levels that this node can use for each resource.Node can use one of the transmitting PSD level for each resource on the uplink.Can by selected by a given resource/the transmitting PSD level that allows defines the use of this resource.In a design, this group launches PSD level can comprise specified PSD level, low PSD level, 0PSD level etc.Specified PSD level in whole available resources can correspond to the maximum transmission power of node.This group transmitting PSD level of node can depend on the power grade of this node.In a design, the transmitting PSDS horizontal group of a given power grade can be that the union of the specified PSD level of the whole power grades less than or equal to this power grade adds 0PSD level.Such as, grand node can comprise the specified PSD level (for grand power grade) of 43dBm, the low PSD level (the specified PSD level corresponding to slight power grade) of 33dBm and 0PSD level.Also the transmitting PSD horizontal group of each power grade can otherwise be defined.

Utility function may be used for calculating the Local Metric for adaptive resource segmentation and overall tolerance.Local Metric and overall tolerance may be used for the performance of a quantification given division of resources.The Local Metric of a given node p can be represented as U (p), and can represent the performance of this node for a given division of resources.The overall tolerance of one group node NS can be represented as V (NS), and can represent the overall performance of this group node for a given division of resources.Local Metric also can be called peer metric, local effectiveness, base station effectiveness etc.Overall tolerance also can be called overall utility, neighborhood effectiveness etc.Can also be whole wireless network calculated population tolerance.The Local Metric that each node can calculate about different possible operations (action) is measured with overall.Can select to make utility function obtain maximum and obtain best overall operation of measuring to use.

In a design, utility function can be defined according to the summation of user rate, as follows:

$U\left(p\right)=\underset{S\left(t\right)=p}{\mathrm{\Σ}}R\left(t\right)$With$V\left(\mathrm{NS}\right)=\underset{p\∈\mathrm{NS}}{\mathrm{\Σ}}U\left(p\right)$Equation (3)

Wherein, R (t) is the speed that UE t realizes.

As shown in equation set (3), Local Metric U (p) of node p can equal the summation of the speed realized by the whole UE served by node p.Overall tolerance V (NS) of adjacent groups NS can equal the summation of the Local Metric of whole node in this adjacent groups.Utility function in equation (3) can not provide Fairness Guarantee.

In another design, utility function can be defined according to minimum user rate, as follows:

$U\left(p\right)=\underset{S\left(t\right)=p}{\min }R\left(t\right)$With$V\left(\mathrm{NS}\right)=\underset{p\∈\mathrm{NS}}{\min }U\left(p\right)$Equation (4)

As shown in equation set (4), Local Metric U (p) of node p can equal the minimum speed limit realized by the whole UE served by node p.Overall tolerance V (NS) of adjacent groups NS can equal the minimum value in the Local Metric of whole node in this adjacent groups.Utility function in equation (4) can guarantee to have the equal grade of service (GoS) for whole UE, can be not too responsive for exceptional value, but can not be provided in trading off between fairness and total throughout.In another design, X% speed utility function can be defined, wherein, Local Metric U (p) of node p can be set as the flank speed of the minimum X% in the whole UE equaling to serve by node p, wherein, X can be the numerical value be applicable to arbitrarily.

In another design again, utility function can be defined according to the summation of the logarithm of user rate, as follows:

$U\left(p\right)=\underset{S\left(t\right)=p}{\mathrm{\Σ}}\log R\left(t\right)$With$V\left(\mathrm{NS}\right)=\underset{p\∈\mathrm{NS}}{\mathrm{\Σ}}U\left(p\right)$Equation (5)

As shown in equation set (5), Local Metric U (p) of node p can equal the summation of the logarithm of the speed of the whole UE served by node p.Overall tolerance V (NS) of adjacent groups NS can equal the summation of the Local Metric of whole node in this adjacent groups.Utility function in equation (5) can provide pro rata equity dispatching.

In another design again, utility function can be defined according to the summation of the logarithm of the logarithm of user rate, as follows:

$U\left(p\right)=\underset{S\left(t\right)=p}{\mathrm{\Σ}}\log \left\{\log R\left(t\right)\right\}$With$V\left(\mathrm{NS}\right)=\underset{p\∈\mathrm{NS}}{\mathrm{\Σ}}U\left(p\right)$Equation (6)

As shown in equation set (6), Local Metric U (p) of node p can equal the summation of the logarithm of the logarithm of the speed of the whole UE served by node p.Overall tolerance V (NS) of adjacent groups NS can equal the summation of the Local Metric of whole node in this adjacent groups.Utility function in equation (6) can illustrate contribution from each UE and more emphasize that afterbody distributes.

In another design again, can according to-1/ (user rate)³the summation of (user rate: user rate) defines utility function, as follows:

$U\left(p\right)=\underset{S\left(t\right)=p}{\mathrm{\Σ}}\frac{-1}{R{\left(t\right)}^3}$With$V\left(\mathrm{NS}\right)=\underset{p\∈\mathrm{NS}}{\mathrm{\Σ}}U\left(p\right)$Equation (7)

As shown in equation set (7), Local Metric U (p) of node p can equal the speed of the whole UE served by node p cube point bear 1 summation.Overall tolerance V (NS) of adjacent groups NS can equal the summation of the Local Metric of whole node in adjacent groups.Utility function in equation (7) can than fair tolerance is more fair in proportion.

Equation set (3) to (7) shows some exemplary design solution of the utility function that can be used for adaptive resource segmentation.Also otherwise utility function can be defined.Can also according to substitution rate or other parameter outside removal rates define utility function.Such as, utility function can be defined according to the function about speed, time delay, queue length etc.

For each design shown in equation set (3) to (7), the Local Metric of this node can be calculated according to the speed of the UE by each node serve.In a design, by supposition, a part (fraction) for each available resources can be distributed to the speed that UE determines each UE.This part can be expressed as α (t, r), and can be regarded as time slice resource r being distributed to UE t wherein.The speed of UE t can be calculated subsequently as follows:

$R\left(t\right)=\underset{r}{\mathrm{\Σ}}\mathrm{\α}(t,r)\·\mathrm{SE}(t,r)\·W\left(r\right)$Equation (8)

Wherein, SE (t, r) is the spectrum efficiency of UE t on resource r, and

W (r) is the bandwidth of resource r.

The spectrum efficiency of UE t on resource r can be determined as follows:

$\mathrm{SE}(t,r)=C\left(\frac{\mathrm{PSD}(p,r)\·G(p,t)}{N_0+\underset{q\≠p}{\mathrm{\Σ}}\mathrm{PSD}(q,r)\·G(q,t)}\right)$Equation (9)

Wherein, PSD (p, r) is the transmitting PSD of service node p on resource r,

PSD (q, r) is the transmitting PSD of adjacent node q on resource r,

G (p, t) is the channel gain between service node p and UE t, and

C () represents capacity function (capacity function).

In equation (9), point expection received power from service node p of subrepresentation at UE t place in bracket.Denominator represents and adds N in the total interference from whole adjacent node at UE t place₀.Service node p transmitting PSD used on resource r and each adjacent node transmitting PSD used on resource r can be known.The channel gain of service node p and adjacent node can be obtained according to the pilot measurements from UE t.Can measure at UE t place/estimate N₀, and comprised in this computation, or wireless network (such as, being reported to service node p) can be reported to by UE t, or can be ignored (such as, when carrying out this calculating at node p place).Capacity function can be constraint capacity function, nothing constraint capacity function or some other functions.

Pre-scheduling device can make spatially maximum, as follows in α (t, r) parameter of utility function:

For 0≤α (t, r)≤1 Hemake U (p) maximum, equation (10) and

U (p)=f ({ R (t) }_{s (t)=p}) equation (11)

Wherein, f () represents the concave function of the speed of the whole UE served by node p.Equation (10) is presented at the convex optimization (convex optimization) in α (t, r) parameter, and can by numerical solution.Pre-scheduling device can be predicted by operation dispatching, and can be different from actual schedule device, and this can make the marginal utility in each scheduling interval (marginal utility) maximum.

The speed of UE t can be retrained as follows:

R (t)≤R_max(t) equation (12)

Wherein, R_maxt () is the maximum rate supported by UE t.

Can global rate R (p) of restraint joint p as follows:

$R\left(p\right)=\underset{S\left(t\right)=p}{\mathrm{\Σ}}R\left(t\right)\≤R_{\mathrm{BH}}\left(p\right)$Equation (13)

Wherein, R_bHp () is the backhaul speed of node p.This backhaul speed can be sent to adjacent node via backhaul, and/or can be sent by aerial, for the decision-making selecting service node for UE.

In a design, adaptive algorithm can be used to adaptive resource segmentation.This algorithm is adaptive, because it can consider current operating situation, described current operating situation may be different for the different piece of wireless network and can change in time.Adaptive algorithm can be performed in a distributed way by each node, and it can attempt making utility function at a group node or likely maximum on whole wireless network.

Fig. 3 shows the design of the process 300 for performing adaptive resource segmentation.Process 300 can be performed by each node in the adjacent groups of Distributed Design scheme.For the sake of simplicity, for node p, process 300 is described below.The Current resource that node p can obtain each node in this adjacent groups uses profile (step 312).For down link, can launch with one group the resource use profile that PSD level defines a node, corresponding one of each available resources launch PSD level.Node p can via backhaul or the Current resource use profile being obtained adjacent node by alternate manner.

Node p can determine the list (step 314) of that can be performed by node p and/or adjacent node, relevant to a division of resources possible operation.The specific resources that each possible operation can correspond to node p uses a specific resources of each adjacent node in profile and this adjacent groups to use profile.Such as, possible operation can make node p change its transmitting PSD in specific resources and/or make adjacent node change its transmitting PSD on the resource.The list of this possible operation can comprise: (i) just can the standard operation of periodic evaluation in without the need to any clearly request situation, and/or (ii) can be evaluated in response to the request from adjacent node operation on demand.The following describe some possible operations.The list of this possible operation can be expressed as A.

Node p can calculate the Local Metric (block 316) about different possible operations.Local Metric can represent the performance of node for a given operation.Such as, the Local Metric based on the utility function in equation (3) can represent the global rate that node p realizes for a specific operation a, and can calculate as follows:

$U(p,a)=\underset{S\left(t\right)=p}{\mathrm{\Σ}}R(t,a)$Equation (14)

Wherein, R (t is a) for operation a, the speed realized in whole available resources by UE t, and

U (p, a) be node p for operation a Local Metric.

Can according to calculate shown in equation (8) and (9) each UE speed R (t, a), wherein, PSD (p, r) can depend on PSD (q, r) that the resource of node p and q uses profile respectively, and be associated with possible operation a.In the design shown in equation (14), first can determine the speed of each UE in whole available resources, subsequently can to the speed summation of the whole UE served by node p with the Local Metric obtaining node p.In another design, first the speed of each UE in each available resources can be determined, next the speed of whole UE in each available resources can be calculated, subsequently can to the speed summation of whole available resources with the Local Metric obtaining node p.Also otherwise can carry out the Local Metric of computing node p for each possible operation, and it can depend on utility function.

The Local Metric about different possible operations can be used to calculate overall tolerance about different possible operations by node p and adjacent node.(p, a), wherein, a ∈ A, is sent to adjacent node (block 318) to the Local Metric U that node p can be calculated.Node p can also from adjacent groups each adjacent node q local tolerance U (q, a), wherein, a ∈ A (block 320).The Local Metric that node p can calculate according to it and the Local Metric received calculate the overall tolerance (block 322) about different possible operations.Such as, the overall tolerance of each possible operation a calculating based on the utility function in equation (3) can be the following is:

$V\left(a\right)=U(p,a)+\underset{q\∈\mathrm{NS}\left(p\right)\backslash \left\{p\right\}}{\mathrm{\Σ}}U(q,a)$Equation (15)

Wherein, V (a) is the overall tolerance about possible operation a.The summation of equation (15) is for except the whole nodes in the external adjacent groups of node p.

After completing metric calculation, node p can select the operation (block 324) with best overall tolerance.Each adjacent node can calculate the overall tolerance about different possible operations similarly, and also can select the operation with best overall tolerance.If node p and adjacent node run in same Local Metric group, they just should select identical operation.Each node can run according to selected operation subsequently, and need not carry out the communication about selected operation each other.But node p and adjacent node thereof can run on different Local Metric, and different the bests can be obtained totally measure.Such as, this can be the situation that node p and adjacent node thereof have different adjacent groups.In the case, node p can consult to determine which carries out operates with adjacent node.This can cause exchanging the overall tolerance about some operations likely among the nodes, and selects to provide the operation of superperformance for node as much as possible.

Select optimum operation howsoever, selected operation all uses profile to be associated with a specific resources of node p.Node p can use profile to use available resources (block 326) according to the resource be associated with selected operation.This resource uses profile can specifically launch PSD horizontal tabulate by one and defines, and wherein corresponding one of each available resources launch PSD level.Node p can use this specifically to launch PSD level to each available resources subsequently.

A large amount of possible operation may be there is for exhaustive search will carry out assessing to find optimum operation.Particularly, if there is the individual possible transmitting PSD level of L for each resource, have K available resources and in adjacent groups, have N number of node, then the total T of possible operation can be given as T=L^kN.Assessing whole T possible operation may be very large in amount of calculation.

The quantity of the possible operation that will assess can be reduced in many ways.In a design, can process each available resources independently, a given operation can change the transmitting PSD of an only resource.The quantity of possible operation can be reduced to T=(L thusⁿ) K.In another design, can be Nx by can adjust its restricted number of launching the node of PSD in given resource for a given operation, it can be less than N.The quantity of possible operation can be reduced to T=(L thus^nx) K.In another design again, the transmitting PSD of a given resource can be increased at every turn or reduce a rank.The quantity of possible operation can be reduced to T=(2 thus^nx) K.Also the quantity of possible operation can be reduced by other simplified way.

In a design, can assess the list of the possible operation that may produce good overall tolerance.The possible operation that good overall tolerance is unlikely provided can be skipped, to reduce computation complexity.Such as, the extra interference making node p and adjacent node increase its transmitting PSD in same resource may to cause on the resource, this can reduce the performance of these two nodes.Therefore this possible operation can be skipped.

Table 2 lists and can carry out for adaptive resource segmentation the dissimilar operation assessed according to a design.

Table 2-action type

Each action type in table 2 can be associated with of a type group possible operation.For each action type only relating to node p, can be K available resources assessment K possible operation.For each action type relating to one or more adjacent node in node p and group Q, can assess multiple possible operation for each available resources, wherein, the quantity of possible operation depends on the size of adjacent groups, the size organizing Q etc.Usually, group Q can comprise one or more adjacent node, and can be restricted to a fractional value (such as, 2 or 3), to reduce the quantity of the possible operation that will assess.

Node p can be that each possible operation of each action type calculates Local Metric.It can be some Local Metrics that the dissimilar operation listed in table 2 calculates by node p that table 3 lists.Local Metric in table 3 is about the different possible operation on a given resource r.This is confined to a resource so that the design reducing computation complexity is consistent with by each possible operation.

Table 3-Local Metric

Can with the Local Metric U with single adjacent node q_0/I(p, q, r), U_0/D(p, q, r), U_i/D(p, q, r) and U_d/I(p, q, r) similar mode defines the Local Metric U of adjacent node group Q respectively_0/I(p, Q, r), U_0/D(p, Q, r), U_i/D(p, Q, r) and U_d/I(p, Q, r).Such as, if its transmitting PSD on resource r is increased one-level, then U by the whole adjacent nodes in group Q_0/I(p, Q, r) can be just the Local Metric of node p.

Node p can according to the following Local Metric calculated about different possible operations: (i) comprises the pilot measurements of the UE of node p in coming comfortable its effective group, and (ii) node p and the resource of adjacent node that is associated with these possible operations use profile.For each possible operation, node p can first such as according to shown in equation (9), calculates the spectrum efficiency SE (t, r) of each UE served by node p on each resource r.The calculating of spectrum efficiency R (t, r) can depend on the scheduling prediction of α (t, the r) value obtaining UE.Can profile be used to obtain PSD (p, r) equation (9) and PSD (q, r) from the resource of node p and q respectively.Can respectively from from UE t, G (p, t) equation (9) and G (q, t) is obtained for the pilot measurements of node p and q.Such as can sue for peace shown in utility function according to the speed in equation (3) subsequently, calculate the Local Metric about this possible operation according to the speed of whole UE in whole available resources.

The pilot measurements of the node be limited in effective group of UE can be utilized to the calculating of Local Metric.Therefore, the precision of Local Metric can be subject to the CoT for selecting the node be included in effective group_minthe impact of threshold value, such as, as shown in equation (1).Higher CoT_minthreshold value can correspond to higher environmental interference amount and lower Local Metric precision.Higher CoT_minthreshold value also corresponds to the looser requirement of UE measurement capability and less effective group.Trading off between complexity can be required according at UE on the one hand, on the other hand according to metric calculation precision, select CoT_minthreshold value.

Node p can exchange (such as, via backhaul) Local Metric with the adjacent node in adjacent groups, can both calculate overall tolerance about different possible operations to make each node.In a design, the Local Metric (the first two Local Metric such as, in table 3) about the possible operation only relating to node p can be sent to the whole adjacent nodes in adjacent groups.Local Metric (four, the centre such as, in table 3 Local Metric) about the possible operation relating to adjacent node q only can be sent to node q.The Local Metric (latter two Local Metric) such as, in table 3 of the possible operation about the adjacent node related in group Q can be sent to each node in group Q.

In a design, periodically can calculate some Local Metrics (such as, front 6 Local Metrics in table 3), and exchange, such as, via standard resource negotiation message between node in adjacent groups.In a design, by message (on-demand message) is asked and exchanges remaining Local Metric on request time (such as, latter two Local Metric and the Local Metric for organizing Q in table 3), can calculate it.Otherwise can calculate Local Metric and exchange among the nodes.

Node p can calculate the Local Metric about different possible operations, and can also from the Local Metric of adjacent node reception about different possible operations.Node p can calculate overall tolerance about different possible operations according to calculated Local Metric and the Local Metric received.Table 4 list can by node p be the dissimilar operation in table 2 calculate some totally measure.

Table 4-totally measures

overall tolerance	explanation
		vc(p, r)	about the overall tolerance of the p-C-r operation on resource r.
vb(p, r)	about the overall tolerance of the p-B-r operation on resource r.
		vr(p, Q, r)	about the overall tolerance of the p-R-r-Q operation on resource r.
vg(p, Q, r)	about the overall tolerance of the p-G-r-Q operation on resource r.
		vcG(p, Q, r)	about the overall tolerance of the p-CG-r-Q operation on resource r.
vbG(p, Q, r)	about the overall tolerance of the p-BG-r-Q operation on resource r.

For the sake of simplicity, below describe and assume such utility function: in this utility function, adjacent groups to equal in this adjacent groups whole node for the summation of the Local Metric of this possible operation for the overall tolerance of a possible operation.Correspondingly can revise overall calculating of measuring for the utility function of other type.Such as, for the utility function in order to make special parameter minimum, the summation of overall tolerance can be replaced with minimum computing.

In a design, the overall tolerance about p-C-r operation can be calculated as follows:

$V_C(p,r)=U_I(p,r)+\underset{q\∈\mathrm{NS}\left(p\right)\backslash \left\{p\right\}}{\mathrm{\Σ}}{U}_{0/I}(q,p,r)$Equation (16), and

Δ V_c(p, r)=V_c(p, r)-V (NS (p)) equation (17)

Wherein, Δ V_c(p, r) is about the change in the overall tolerance of p-C-r operation, and

V (NS (p)) is the overall tolerance that the Current resource of adjacent groups uses.

As shown in equation (16), can according to the Local Metric U calculated by node p_i(p, r) and the Local Metric U received from adjacent node_0/I(q, p, r) carrys out calculated population tolerance V_c(p, r).As shown in equation (17), can calculate and use the change in overall tolerance, replacing the absolute value of equation (16).

In a design, the overall tolerance about p-B-r operation can be calculated as follows:

$V_B(p,r)=U_D(p,r)+\underset{q\∈\mathrm{NS}\left(p\right)\backslash \left\{p\right\}}{\mathrm{\Σ}}{U}_{0/D}(q,p,r)$Equation (18) and

Δ V_b(p, r)=V_b(p, r)-V (NS (p)) equation (19)

Wherein, Δ V_b(p, r) is about the change in the overall tolerance of p-B-r operation.

As shown in equation (18), can according to the Local Metric U calculated by node p_d(p, r) and the Local Metric U received from adjacent node_0/D(q, p, r) carrys out calculated population tolerance V_b(p, r).Node p can exchange with adjacent node and totally measure V_c(p, r) and V_b(p, r) (or corresponding Δ V_c(p, r) and Δ V_b(p, r)), to use when calculating other and totally measuring.

In a design, the overall tolerance about p-G-r-Q operation can be calculated as follows.First, the initial estimate that this is totally measured can be calculated as follows:

$V_{G,0}(p,Q,r)=U_{0,I}(p,Q,r)+\underset{q\∈Q}{\mathrm{\Σ}}\{V_C(q,r)-U_{0/I}(p,q,r)\}$Equation (20), and

${\mathrm{\ΔV}}_{G,0}(p,Q,r)=V_{G,0}(p,Q,r)-U\left(p\right)-\underset{q\∈Q}{\mathrm{\Σ}}\{V\left(\mathrm{NS}\left(q\right)\right)-U\left(p\right)\}$Equation (21)

Wherein, U (p) is the Local Metric that node p uses for Current resource,

V_{g, 0}(p, Q, r) is the initial estimate of the overall tolerance about p-G-r-Q operation, and

Δ V_{g, 0}(p, Q, r) is the initial estimate to the change in this totally tolerance.

As shown in equation (20), can according to the Local Metric U calculated by node p_0/I(p, q, r) and U_0/I(p, Q, r) and the overall tolerance V received from adjacent node_c(q, r) calculates V_{g, 0}(p, Q, r).If this initial estimate seems to be hopeful (such as, if the change in this overall tolerance is greater than threshold value), then can calculate this as follows more accurately and totally measure:

$V_G(p,Q,r)=\underset{n\∈\mathrm{NS}\left(p\right)}{\mathrm{\Σ}}{U}_{0/I}(n,Q,r)+\underset{q\∈Q}{\mathrm{\Σ}}(V_C(q,r)-\underset{n\∈N1}{\mathrm{\Σ}}{U}_{0/I}(n,q,r))$Equation (22)

${\mathrm{\ΔV}}_G(p,Q,r)=V_G(p,Q,r)-V\left(\mathrm{NS}\left(p\right)\right)-\underset{q\∈Q}{\mathrm{\Σ}}(V\left(\mathrm{NS}\left(q\right)\right)-\underset{n\∈N1}{\mathrm{\Σ}}U\left(n\right))$Equation (23)

Wherein, Δ V_g(p, Q, r) is about the change in the overall tolerance of p-G-r-Q operation, and

N1＝NS(p)∩NS(q)。

In a design, only under initial estimate seems situation likely, node p is just from the Local Metric U adjacent node request equation (22)_0/I(n, q, r) and U_0/I(n, Q, r).This design can be reduced to adaptive resource segmentation and carry out the amount of information that exchanges via backhaul.

In a design, can calculate in the mode similar with the overall tolerance operated about p-G-r-Q the overall tolerance operated about p-R-r-Q.Equation (18) to (21) may be used for calculating the overall tolerance about p-R-r-Q operation, although use Local Metric U respectively_0/D(p, q, r), U_0/D(p, Q, r), U_0/D(n, q, r) and U_0/D(n, Q, r) replaces Local Metric U_0/I(p, q, r), U_0/I(p, Q, r), U_0/I(n, q, r) and U_0/I(n, Q, r).

In a design, the overall tolerance about p-BG-r-Q operation can be calculated as follows.First, can as follows calculated population tolerance initial estimate:

$\begin{array}{c}{V}_{\mathrm{BG},0}(p,Q,r)=U_{D/I}(p,Q,r)+\underset{n\∈N2}{\mathrm{\Σ}}{U}_{0/D}(n,p,r)+\\ \underset{q\∈Q}{\mathrm{\Σ}}\{U_{I/D}(q,p,r)+V_C(q,r)-U_I(q,r)-U_{0/I}(p,q,r)\}\end{array},$Equation (24)

$\begin{array}{c}{\mathrm{\ΔV}}_{\mathrm{BG},0}(p,Q,r)=V_{\mathrm{BG},0}(p,Q,r)-V\left(\mathrm{NS}\left(p\right)\right)-\\ \underset{q\∈Q}{\mathrm{\Σ}}\{V\left(\mathrm{NS}\left(q\right)\right)-U\left(p\right)-U\left(q\right)\}\end{array}$Equation (25)

Wherein, V_{bG, 0}(p, Q, r) is the initial estimate of the overall tolerance about p-BG-r-Q operation,

Δ V_{bG, 0}(p, Q, r) is the initial estimate of the change during this is totally measured, and

N2＝NS(p)\(Q∪{p})。

As shown in equation (24), V can be calculated according to following_{bG, 0}(p, Q, r): the Local Metric U that (i) is calculated by node p_0/I(p, q, r) and U_d/I(p, Q, r), and the Local Metric U that (ii) receives from adjacent node_i(q, r), U_0/D(n, p, r) and U_i/D(q, p, r) and totally measure V_c(q, r).If this initial estimate seems to be hopeful, then can calculate this as follows more accurately and totally measure:

$V_{\mathrm{BG}}(p,Q,r)=\underset{n\∈\mathrm{NS}\left(p\right)}{\mathrm{\Σ}}{U}_{0/D/I}(n,p,Q,r)+\underset{q\∈Q}{\mathrm{\Σ}}(V_C(q,r)-\underset{n\∈N1}{\mathrm{\Σ}}{U}_{0/I}(n,q,r))$Equation (26)

${\mathrm{\ΔV}}_{\mathrm{BG}}(p,Q,r)=V_{\mathrm{BG}}(p,Q,r)-V\left(\mathrm{NS}\left(p\right)\right)-\underset{q\∈Q}{\mathrm{\Σ}}(V\left(\mathrm{NS}\left(q\right)\right)-\underset{n\∈N1}{\mathrm{\Σ}}U\left(n\right))$Equation (27)

Wherein, Δ V_bG(p, Q, r) is about the change in the overall tolerance of p-BG-r-Q operation.If this initial estimate seems to be hopeful, then node p just can from the Local Metric U adjacent node request equation (26)_0/I(n, q, r) and U_0/D/I(n, p, Q, r).

In a design, can calculate in the mode similar with the overall tolerance operated about p-BG-r-Q the overall tolerance operated about p-CR-r-Q.Equation (24) to (27) may be used for calculating the overall tolerance about p-CR-r-Q operation, such as, uses U respectively_0/D(n, q, r) and U_0/I/D(n, p, Q, r) replaces the Local Metric U in equation (26)_0/I(n, q, r) and U_0/D/I(n, p, Q, r).

Equation (16) to (27) shows the example calculation for the overall tolerance in table 4, and these are totally measured is about the dissimilar operation in table 2.Some totally tolerance can only calculate according to Local Metric, such as, shown in equation (16) and (18).Some other totally tolerance can calculate according to the combination of Local Metric and overall tolerance, such as, shown in equation (22) and (26).Use some overall tolerance to calculate other totally to measure and can simplify calculating.Usually, only calculated population tolerance can be carried out according to Local Metric or according to Local Metric and other overall measurement.Message can be taken turns or take turns more to node to exchange Local Metric and/or totally to measure by means of one.

Otherwise can also carry out calculated population tolerance, such as, according to other equation, other Local Metric etc.Usually, any action type group can be supported.The overall tolerance about supported action type can be calculated, and overall tolerance can be defined in many ways.

Simulate the adaptive resource segmentation of the small-scale wireless network of the node for having two power grades.In this simulation, adjacent groups comprises two nodes (or grand node) for macro base station and six nodes (or pico node) for femto base station.Each grand node has the specified PSD level (being expressed as 2) of three PSD level-43dBm, the low PSD level (being expressed as 1) of 33dBm, and 0PSD level (being expressed as 0).Each pico node has specified PSD level (being expressed as 1) and the 0PSD level (being expressed as 0) of two PSD level-33dBm.Always have four resources to can be used for splitting between these nodes.Always have 16 UE to be distributed in whole wireless network.

Fig. 4 shows wireless network in this simulation.Two grand nodes are expressed as M1 and M2, and four pico node are expressed as P1 to P4, and 16 UE are expressed as UE1 to UE16.Fig. 4 also show the result split according to the adaptive resource of above-mentioned adaptive algorithm.Being close to each node is one group of four numeral, and it represents the transmitting PSD level of this node in four available resources.Such as, grand node M 2 is associated with " 0211 ", and it is meant to that 0 transmitting PSD is used for resource 1,43dBm and is used for resource 2,33dBm for resource 3,33dBm for resource 4.

Fig. 4 also show the communication link between each UE and service node thereof.With the communication link of two each UE of figure notation.Numeral above distributes to total part (fraction) of the resource of this UE.Total speed R (t) that beneath numeral is realized by this UE.Such as, the communication link from UE9 to grand node M 2 represents distributes to UE 9 by 2.2 average 3 resources, and it realizes the speed of 3.9Mbps.For each node, the resource summation distributed to by whole UE of this node serve should equal to be distributed to by adaptive resource the resource of this node.

Table 5 lists the performance of adaptive resource segmentation and the performance of multiple fixed resource splitting scheme.Split for fixing X:Y, give grand node by X Resourse Distribute, by Y Resourse Distribute to pico node, each node uses specified PSD level in each resource distributing to this node, wherein, for the example shown in Fig. 4, and X+Y=4.For adaptive resource segmentation, can be the resource of each peer distribution configurable number, the resource that each grand node can distribute at each is launched with 43dBm or 33dBm.

Table 5 shows three overall tolerance of different resource splitting scheme.It is utility function based on shown in equation (6) that log log IU totally measures.Minimum-rate is totally measured (Rmin) is based on the utility function shown in equation (4).It is based on the utility function shown in equation (3) that aggregate rate type totally measures (Rsum).As shown in table 5, adaptive resource segmentation can provide performance more better than fixed resource splitting scheme.

Table 5

Division of resources scheme	log log IU	Rmin	Rsum	Unit
					Adaptive resource is split	6.37	3.29	119.64	Mbps
Fixing 1:3 segmentation	4.85	1.73	92.81	Mbps
					Fixing 2:2 segmentation	4.23	1.15	87.56	Mbps
Fixing 3:1 segmentation	2.72	0.58	82.33	Mbps

In a design, adaptive resource segmentation can be performed for whole resources of the transmission that can be used in wireless network.In another design, adaptive resource segmentation can be performed for the subset of available resources.Such as, can split according to fixed resource, be the first subset of grand peer distribution resource, is the second subset of pico node Resources allocation.Can split according to adaptive resource, remaining available resources are dynamically distributed to grand node or pico node.For the example shown in Fig. 4, give grand node, by a Resourse Distribute to pico node by a Resourse Distribute; According to adaptive resource segmentation, remaining two resource dynamic are distributed to grand node or pico node.This design can reduce computational complexity.

For the sake of simplicity, the adaptive resource the foregoing described for down link is split.The adaptive resource that can perform in a similar fashion for up link is split.In a design, in the mode similar with the PSD horizontal group for down link, one group of target jamming thermal noise ratio (interference-over-thermal, IoT) level can be used for division of resources on uplink.It can be each resource selection target IoT level in up link, and the transmission that can control from each UE in each resource, with the target IoT level making each the adjacent node place actual IoT on the resource in effective group of this UE be equal to or less than this resource at this adjacent node place.Utility function can be defined to quantize data transmission performance on uplink, and utility function can be the function of user rate summation, the minimum value of user rate function, etc.Each UE speed on uplink can be the function of transmitting power, channel gain and target IoT level etc.The Local Metric that can calculate about different possible operations according to utility function is measured with overall.Each possible operation can be associated with the list of the target IoT level for whole available resources of each node in adjacent groups.The possible operation with best overall tolerance can be selected to use.

Fig. 5 shows the design for supporting the process 500 communicated.Process 500 can perform by node (as described below) or by some other entities (such as, network controller).Node can be base station, relaying or some other entities.Node can obtain the overall tolerance about the multiple possible operations relevant to division of resources, available resources to be distributed to the group node (block 512) comprising this node.Each possible operation can use profile to be associated with of this group node group resource, and the corresponding resource of each node uses profile.Each resource uses profile can represent the use that be allowed to of this specific node to available resources.This node can be determined according to the overall tolerance about the plurality of possible operation to distribute (block 514) for the available resources of this group node.

Described available resources can be for chronomere, cps, T/F unit etc.In a design, available resources may be used for down link.In this design, a group of can be allowed to use with this node of each node in this group node is launched PSD level and is associated.Each resource uses profile can comprise the list of the transmitting PSD level for these available resources, and corresponding one of each available resources launch PSD level.Transmitting PSD level for each available resources can be one that this group launches in PSD level.In another design, available resources may be used for up link.In this design, each resource uses profile can comprise the list of the target IoT level for these available resources, the corresponding target IoT level of each available resources.

In a design of block 514, this node can select one of the plurality of possible operation according to the overall tolerance about the plurality of possible operation.This node can according to be associated with selected operation and the resource that can be used for this node uses profile to determine the resource distributing to this node.Node can use profile to come to be at least one UE schedule data transmission in available resources according to the resource for this node.

Fig. 6 shows the design for supporting the device 600 communicated.Device 600 comprises: module 612, in order to obtain the overall tolerance about the multiple possible operations relevant to division of resources, available resources are distributed to a group node; And module 614, in order to determine to distribute for the available resources of this group node according to the overall tolerance about the plurality of possible operation.

Fig. 7 shows the design of the process 700 for performing adaptive resource segmentation, and it may be used for the block 512 and 514 in Fig. 5.Node can calculate the Local Metric about the multiple possible operations relevant to division of resources, available resources to be distributed to the group node (block 712) comprising this node.Calculated Local Metric can be sent at least one adjacent node in this group node by this node, can calculate overall tolerance (block 714) about the plurality of possible operation to make this adjacent node.This node can from the Local Metric (block 716) of this at least one adjacent node reception about the plurality of possible operation.This node can according to the overall tolerance (block 718) determined about the Local Metric calculated of these possible operations and the Local Metric that receives about the plurality of possible operation.Local Metric about a possible operation can represent the performance that this node realizes for this possible operation.Overall tolerance about a possible operation can represent the overall performance that this group node realizes for this possible operation.

This node can select one of the plurality of possible operation according to the overall tolerance about the plurality of possible operation, such as, selects the operation (block 720) with best overall tolerance.This node can according to be associated with selected operation and the resource that may be used for this node uses profile to use available resources (block 722).

In a design of block 712, for each possible operation, this node can determine at least one speed with at least one UE of this node communication according to following: the resource that (i) is associated with this possible operation uses profile group, and (ii) channel gain between each UE and this node and adjacent node.The Local Metric that this node can be determined about this possible operation according to this at least one speed subsequently.The Local Metric about the plurality of possible operation can be calculated according to the function about speed or time delay or queue size or some other parameters or its combination.The Local Metric about the plurality of possible operation can also be calculated according to the function about speed summation or speed minimum value or the summation of value determined based on speed etc.

In a design of block 714 and 716, can the first subset of Local Metric of receiving of periodically clearing house calculates between this node and this at least one adjacent node the first subset sums of Local Metric.Can when there being request between this node and this at least one adjacent node the second subset of the Local Metric that the second subset sums of the Local Metric that clearing house calculates receives.

In a design of block 718, for each possible operation, this node can by by this node calculate, about this possible operation Local Metric with to receive from this at least one adjacent node, combine about at least one Local Metric of this possible operation, to obtain the overall tolerance about this possible operation.

In a design, each in the plurality of possible operation only can affect one in available resources.In another design, the transmitting PSD (or target IoT) of any given node in this group node can be changed maximum one-level by each possible operation.In a design, one group of action type such as shown in table 2 can be supported.Each in the plurality of possible operation can be one in this group action type.The plurality of possible operation can comprise (i) first possible operation, this node increases it and launches PSD, (ii) the second possible operation, this node reduces it and launches PSD, (iii) the 3rd possible operation, one or more adjacent node increases it and launches PSD, (iv) the 4th possible operation, one or more adjacent node reduces it and launches PSD, (v) the 5th possible operation, this node increases its transmitting PSD and one or more adjacent node reduces its transmitting PSD, (vi) the 6th possible operation, this node reduces its transmitting PSD and one or more adjacent node increases its transmitting PSD, or (vii) combination of above-mentioned possible operation.

In a design, each UE can be associated with effective group of having higher than the received signal quality of threshold value or the node of received signal strength.Can determine this group node according to effective group of UE, and it can comprise the node in effective group of the UE of (i) and this node communication, and/or (ii) provides the node of service for the UE with effective group that comprises this node.In a design, this group node can comprise the node of different capacity grade.Such as, this group can comprise the first node with the first maximum transmit power level and the Section Point with second/different maximum transmit power level.In another design, this group node can comprise the node of equal-wattage grade.

More than describe for Distributed Design scheme, in described Distributed Design scheme, the node in node group each can for calculates and exchanges about different possible operations Local Metric and totally measure.For centralized design, the Local Metric that designated entities can calculate about different possible operations is measured with overall, and can select optimum operation.

Fig. 8 shows for having the design of process 800 of communication in wireless networks of adaptive resource segmentation.Process 800 can perform by UE (as described below) or by some other entities.UE can carry out pilot measurement (block 812) to the node that can be detected by this UE.Pilot measurement may be used for effective group that determines this UE.Pilot measurement can also be used for calculating the Local Metric for adaptive resource segmentation.

UE can from the distribution (block 814) of node reception at least one resource.Adaptive resource segmentation can be performed available resources to be distributed to the group node comprising this node.This node can be given by the subset allocation of available resources by means of adaptive resource segmentation.This at least one resource being assigned to UE can come from this subset of the available resources distributing to this node.

UE can carry out communicate (block 816) with this node in this at least one resource.In a design of block 816, UE can receive transfer of data from this node in this at least one resource.The transmitting PSD level that can be allowed to use in this at least one resource in each resource, with this node on the resource by this node is to send transfer of data.In another design of block 816, UE can send transfer of data to this node in this at least one resource.Can by UE in this at least one resource in each resource, with based at least one adjacent node at least one target IoT level on the resource and the transmit power level determined sends transfer of data.

Fig. 9 shows for having the design of device 900 of communication in wireless networks of adaptive resource segmentation.Device 900 comprises: module 912, for carrying out pilot measurement to the node that can be detected by UE; Module 914, for receiving the distribution at least one resource at UE place from node; And module 916, for being communicated with this node in this at least one resource by this UE.

Module in Fig. 6 and 9 can comprise processor, electronic device, hardware device, electronic building brick, logical circuit, memory, software code, firmware code etc., or its combination.

Figure 10 shows the block diagram of the design of base station/node 110 and UE 120, and it can be a base station in Fig. 1 and a UE.Base station 110 can be furnished with T antenna 1034a to 1034t, and UE 120 can be furnished with R antenna 1052a to 1052r, wherein usually, and T >=1 and R >=1.

At base station 110 place, launching processor 1020 can receive from data source 1012 data being used for one or more UE, and from controller/processor 1040 receiving control information.Processor 1020 can process (such as, coding, intertexture and modulation) data and control information, to obtain data symbol and control character respectively.Processor 1020 can also be that pilot tone or reference signal produce frequency pilot sign.Launch (TX) multiple-input and multiple-output (MIMO) processor 1030 and can perform spatial manipulation (such as on data symbol, control character and/or frequency pilot sign, precoding) (if it is available), and T output symbol stream can be provided to T modulator (MOD) 1032a to 1032t.Each modulator 1032 can process respective output symbol stream (such as, using OFDM etc.), to obtain output sample stream.Each modulator 1032 can process further (such as, being converted to analog signal, amplification, filtering and up-conversion) output sample stream is to obtain down link signal.T the down link signal from modulator 1032a to 1032t can be launched respectively via T antenna 1034a to 1034t.

At UE 120 place, the signal of reception from base station 110 receiving downlink signal, and can be supplied to demodulator (DEMOD) 1054a to 1054r by antenna 1052a to 1052r respectively.Each demodulator 1054 can regulate (such as, filtering, amplification, down-conversion and digitlization) its Received signal strength to obtain input amendment.Each demodulator 1054 can process input amendment (such as, with OFDM etc.) further to obtain receiving symbol.MIMO detector 1056 can obtain receiving symbol from whole R demodulator 1054a to 1054r, detects (if it is available), and provide detected symbol to reception semiology analysis MIMO.The data of the decoding being used for UE 120 are supplied to data sink 1060, and the control information of decoding are supplied to controller/processor 1080 by detected symbol that receiving processor 1058 can process (such as, demodulation, deinterleaving and decoding).

On uplink, at UE 120 place, launch processor 1064 and can receive and the data processed from data source 1062 and the control information carrying out self-controller/processor 1080.Processor 1064 can also be that pilot tone or reference signal produce frequency pilot sign.The symbol carrying out spontaneous emission processor 1064 can carry out precoding (if it is available) by TX MIMO processor 1066, be further processed (such as by modulator 1054a to 1054r again, for SC-FDM, OFDM etc.), and be sent to base station 110.At base station 110 place, uplink signal from UE 120 can be received by antenna 1034, processed by demodulator 1032, detected (if it is available) by MIMO detector 1036, and processed further with the data obtaining the decoding sent by UE 120 and control information by receiving processor 1038.The data of decoding can be supplied to data sink 1039 by processor 1038, and the control information of decoding is supplied to controller/processor 1040.

Controller/processor 1040 and 1080 can instruct the operation in base station 110 and UE 120 place respectively.Channel processor 1084 can carry out pilot measurement, and pilot measurement may be used for effective group and calculating channel gain, speed, tolerance etc. determining UE 120.Processor 1040 and/or other processors at base station 110 place and module can perform or the process 300 in guidance diagram 3, the process 500 in Fig. 5, the process 700 in Fig. 7 and/or other processes for technology described herein.Processor 1080 and/or other processors at UE 120 place and module can perform or the process 800 in guidance diagram 8 and/or other processes for technology described herein.Memory 1042 and 1082 can be respectively base station 110 and UE 120 stores data and program code.Scheduler 1044 can be data transmission scheduling UE on the downlink and/or uplink.

It will be appreciated by those skilled in the art that and arbitrary multiple different technology and technique can be used to represent information and signal.Such as, the data mentioned in the above description, instruction, order, information, signal, bit, symbol and chip can represent with voltage, electric current, electromagnetic wave, magnetic field or magnetic particle, light field or optical particle or its combination in any.

Those skilled in the art be to be further appreciated that, all can be implemented as electronic hardware, computer software or its combination in conjunction with disclosure describes various exemplary logical block, module, circuit and algorithm steps herein.In order to clearly represent this interchangeability between hardware and software, in its function aspects, describe, in general terms is carried out to various exemplary assembly, block, module, circuit and step above.Be embodied as hardware as this function or be embodied as software, the design constraint depending on specific application and whole system is applied.Those skilled in the art for each application-specific, can realize described function in the mode of accommodation, but, decision-making should do not realized be interpreted as this deviating from scope of the present invention.

By general processor, digital signal processor (DSP), application-specific integrated circuit (ASIC) (ASIC), field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components or its combination in any performing function described herein can be designed to, realize or perform in conjunction with various exemplary logical block, module and the circuit described by disclosure file.General processor can be microprocessor, but alternatively, this processor also can be the processor of any routine, controller, microcontroller or state machine.Processor also can be implemented as the combination of calculating device, such as, and the combination of the combination of DSP and microprocessor, the combination of multi-microprocessor, one or more microprocessor and DSP kernel or other this kind of structure any.

Step in conjunction with the method described by disclosure file or algorithm directly can be presented as hardware, the software module performed by processor or the combination of the two.Software module can be arranged in the storage medium of RAM memory, flash memory, ROM memory, eprom memory, eeprom memory, register, hard disk, removable dish, CD-ROM or other form any well known in the art.A kind of exemplary storage medium can be coupled to processor, makes processor can from this read information and can to this storage medium written information.Alternatively, storage medium can be integrated in processor.Processor and storage medium can be arranged in ASIC.ASIC can be arranged in user terminal.Alternatively, processor and memory can be arranged in user terminal as discrete assembly.

In one or more exemplary design solution, described function can realize in hardware, software, firmware or its combination in any.If realized in software, then described function can as one or more instruction or code storage on a computer-readable medium, or be sent by computer-readable medium.Computer-readable medium comprises computer-readable storage medium and communication media, and communication media comprises the arbitrary medium being convenient to transmit computer program from a position to another location.Storage medium can be any usable medium can accessed by all-purpose computer or special-purpose computer.Exemplarily and without limitation, this computer-readable medium can comprise RAM, ROM, EEPROM, CD-ROM or other disk storage, magnetic disc store or other magnetic storage apparatus or can be used for the form carrying of instruction or data structure or other medium any storing the prediction program code module can accessed by all-purpose computer or special-purpose computer or general processor or application specific processor.In addition, suitably computer-readable medium is called by connecting arbitrarily.Such as, if use coaxial cable, optical cable, twisted-pair feeder, digital subscriber line (DSL) or such as infrared, radio and microwave wireless technology by software from website, server or other remote source send, then the wireless technology of coaxial cable, optical cable, twisted-pair feeder, DSL or such as infrared, radio and microwave is included in the definition of medium.Disk used herein and disc comprise the disc that compacts (CD), laser disc, CD, digital multi disc (DVD), floppy disk and blue light disk, wherein disk is usually with magnetic means rendering data, and disc is by laser rendering data to be optically.The combination of above-mentioned medium is also included within the scope of computer-readable medium.

The above description of disclosure is provided, can realizes to make those skilled in the art or use present disclosure.Those skilled in the art easily will know the various amendments to these disclosures, and when not departing from the spirit or scope of present disclosure, General Principle defined herein can be applied to other change.Therefore, disclosure file is not intended to be limited to example as herein described and design, and should be given the maximum magnitude consistent with principle disclosed herein and novel feature.

Claims (27)

1., for a method for radio communication, comprising:

The effective node group retained for a described UE is determined at the service node place for first user device (UE);

Determine adjacent node group, wherein, described adjacent node group comprises at least one node except described service node, wherein, the terminal that at least one node described is being served to described service node is signaled or causes interference to it, or serves at least one UE being signaled by the transmission from described service node or disturbed;

At least in part based on the realized user transmission rate of first group of UE and the one or more performance metrics from least one node described in described adjacent node group that comprise a described UE, determine that the resource of the described service node of serving described first group of UE uses profile, wherein, described one or more performance metric is associated with the second group of UE served by least one node described; And

Use profile based on described resource, at least one transfer of data is sent to the one or more UE in described first group of UE.

2. the method for claim 1, wherein determine described effective group to comprise further:

Exceed carrier to thermal noise ratio (CoT) threshold value in response to the pilot measurements for a node, this node is included in described effective node group.

3. the method for claim 1, wherein determine that described resource uses profile to comprise:

User's transmission rate can be realized at least in part described in described first group of UE, assess utility function;

Described one or more performance metric is received from the one or more nodes described adjacent node group, wherein, described one or more performance metric is at least in part based on the described utility function assessed for each group of UE served by described one or more node.

4. method as claimed in claim 3, wherein, described utility function is based on the described summation realizing user's transmission rate.

5. method as claimed in claim 3, also comprises:

Be based upon the spectrum efficiency for each available resources described in described one group of available resources of each UE in the estimated part of each available resources in one group of available resources that described first group of UE distribute and described first group of UE, calculate and describedly can realize user's transmission rate.

6. method as claimed in claim 5, also comprises:

At least in part based on the power spectral density (PSD) of described service node in each available resources described in described one group of available resources, calculate the described spectrum efficiency for each available resources described in described one group of available resources of described each UE.

7. method as claimed in claim 3, also comprises:

Use pre-scheduling device make described utility function distribute in part estimated by each available resources of described first group of UE maximum.

8. method as claimed in claim 3, wherein, determine that described resource uses profile to comprise:

Obtain the described one or more performance metric for multiple division of resources action; And

Assess the described one or more performance metric for described multiple division of resources action.

9. method as claimed in claim 3, wherein, determine that described resource uses profile to comprise:

Obtain the described one or more performance metric at least one division of resources action in multiple division of resources action, described division of resources action comprises following one or more: the transmit power spectral density (PSD) increasing resource, reduce the described transmitting PSD of described resource, to resource described in the node request in described adjacent node group, the described node in described adjacent node group is asked to reduce the described transmitting PSD in described resource, the described transmitting PSD in described resource is increased to the described node instruction in described adjacent node group, described resource is permitted to the described node in described adjacent node group, or the combination of above-mentioned action.

10. method as claimed in claim 3, wherein, described utility function is based on the described minimum user's transmission rate realized in user's transmission rate.

11. methods as claimed in claim 3, wherein, described utility function is based on following at least one item: the summation of the logarithm of the described user's of realization transmission rate, described realize the logarithm of the logarithm of user's transmission rate summation, or more every combination.

12. methods as claimed in claim 3, wherein, described utility function based on described realize user's transmission rate cube the summation of negative inverse.

13. methods as claimed in claim 5, also comprise:

Based on the bandwidth of each resource in described one group of available resources, calculate the described of each UE described in described first group of UE and realize user's transmission rate.

14. methods as claimed in claim 6, also comprise:

At least in part based on the power spectral density (PSD) of adjacent node in each resource described in described one group of available resources, calculate the described spectrum efficiency for each resource described in described one group of available resources of described each UE.

15. methods as claimed in claim 6, also comprise:

At least in part based on the channel gain between described service node and described each UE, calculate the described spectrum efficiency for each resource described in described one group of available resources of described each UE.

16. methods as claimed in claim 6, also comprise:

At least in part based on capacity function, calculate the described spectrum efficiency for each resource described in described one group of available resources of described each UE.

17. 1 kinds, for the device of radio communication, comprising:

For determining the module of the effective node group retained for a described UE at the service node place for first user device (UE);

For determining the module of adjacent node group, wherein, described adjacent node group comprises at least one node except described service node, wherein, the terminal that at least one node described is being served to described service node is signaled or causes interference to it, or serves at least one UE being signaled by the transmission from described service node or disturbed;

For at least in part based on the realized user transmission rate of first group of UE and the one or more performance metrics from least one node described in described adjacent node group that comprise a described UE, determine that the resource of the described service node of serving described first group of UE uses the module of profile, wherein, described one or more performance metric is associated with the second group of UE served by least one node described; And

For using profile based on described resource, at least one transfer of data is sent to the module of the one or more UE in described first group of UE.

18. devices as claimed in claim 17, wherein, described for determining that the module of described effective group comprises further:

For exceeding carrier to thermal noise ratio (CoT) threshold value in response to the pilot measurements for a node, this node is included in the module in described effective node group.

19. devices as claimed in claim 17, wherein, for determining that described resource uses the module of profile to comprise:

For user's transmission rate can be realized at least in part described in described first group of UE, assess the module of utility function;

For receiving the module of described one or more performance metric from the described one or more nodes in described adjacent node group, wherein, described one or more performance metric is at least in part based on the described utility function assessed for each group of UE served by described one or more node.

20. devices as claimed in claim 19, wherein, described utility function is also based on following at least one item: the summation of the logarithm of the minimum user rate in the summation of the described user's of realization transmission rate, the described user's of realization transmission rate, the summation of the logarithm of the described user's of realization transmission rate, the logarithm of the described user's of realization transmission rate or described realize user's transmission rate cube negative summation reciprocal, or more every combination.

21. devices as claimed in claim 19, comprise further:

For calculating the described module that can realize user's transmission rate based on following at least one item at least in part: distribute to the bandwidth of each available resources described in the spectrum efficiency for each available resources described in described one group of available resources of each UE in the estimated part of each resource in one group of available resources of described first group of UE, described first group of UE or described one group of available resources, or more every combination.

22. devices as claimed in claim 21, comprise further:

For at least in part based on the following one or more module calculating the spectrum efficiency for each available resources described in described one group of available resources of described each UE: the power spectral density of described service node in each available resources (PSD), the adjacent node PSD in each available resources described in described one group of available resources, channel gain between described service node and described each UE, capacity function, or more every combination.

23. devices as claimed in claim 19, comprise further:

Module maximum in part estimated by each available resources of described first group of UE is being distributed to make described utility function for using pre-scheduling device.

24. devices as claimed in claim 19, wherein, described for determining that described resource uses the module of profile to comprise:

For obtaining the module of the described one or more performance metric for multiple division of resources action; And

For assessment of the module of the described one or more performance metric for described multiple division of resources action.

25. devices as claimed in claim 19, wherein, described for determining that described resource uses the module of profile:

Obtain the described one or more performance metric at least one division of resources action in multiple division of resources action, described division of resources action comprises following one or more: the transmit power spectral density (PSD) increasing resource, reduce the described transmitting PSD of described resource, to resource described in the node request in described adjacent node group, the described node in described adjacent node group is asked to reduce the described transmitting PSD in described resource, the described transmitting PSD in described resource is increased to the described node instruction in described adjacent node group, described resource is permitted to the described node in described adjacent node group, or the combination of above-mentioned action.

26. 1 kinds of computer programs, comprising:

Non-volatile computer-readable medium, comprising:

For determining the code of the effective node group retained for a described UE at the service node place for first user device (UE);

For determining the code of adjacent node group, wherein, described adjacent node group comprises at least one node except described service node, wherein, the terminal that at least one node described is being served to described service node is signaled or causes interference to it, or serves at least one UE being signaled by the transmission from described service node or disturbed;

For at least in part based on the realized user transmission rate of first group of UE and the one or more performance metrics from least one node in described adjacent node group that comprise a described UE, determine that the resource of the described service node of serving described first group of UE uses the code of profile, wherein, described one or more performance metric is associated with the second group of UE served by least one node described; And

For using profile based on described resource, at least one transfer of data is sent to the code of the one or more UE in described first group of UE.

27. 1 kinds, for the device of radio communication, comprising:

Memory;

At least one processor, is coupled to described memory and is configured to:

The effective node group retained for a described UE is determined at the service node place for first user device (UE);

Determine adjacent node group, wherein, described adjacent node group comprises at least one node except described service node, wherein, the terminal that at least one node described is being served to described service node is signaled or causes interference to it, or serves at least one UE being signaled by the transmission from described service node or disturbed;

At least in part based on the realized user transmission rate of first group of UE and the one or more performance metrics from least one node in described adjacent node group that comprise a described UE, determine that the resource of the described service node of serving described first group of UE uses profile, wherein, described one or more performance metric is associated with the second group of UE served by least one node described; And

Use profile based on described resource, at least one transfer of data is sent to the one or more UE in described first group of UE.